IRC Logs for #crux Wednesday, 2012-08-22

*** mike_k_ has joined #crux00:43
*** lasso|qt has joined #crux01:07
*** nogagplz_ has joined #crux01:09
*** SiFuh has quit IRC02:15
*** v33 has quit IRC02:29
*** sepen has joined #crux02:45
*** SiFuh has joined #crux03:01
Romsterv33, there is freetype-3203:18
Romsteri don't care that v33 is gone he should look in the irc log03:18
frinnsthe wont03:56
sepenoops seems that new chromium update failed to load some icons used on 'Downloads window', someone experienced the same behaviour?03:57
teK_I must have missed that one (and I switched ahead to 22.x)03:58
sepen21 is marked as stable03:58
teK_I wanted to see if the 22.x builtin flash did suck less03:58
teK_I doesn't.03:58
sepenthey update again the stable branch03:59
sepento 21.0.1180.8103:59
teK_I have to make a benchmark 2x XEONs Quad i7 with regard to compile performance; test-target: chromium :P04:00
*** sepen has quit IRC06:13
*** acrux_ has joined #crux06:26
*** acrux has quit IRC06:28
*** acrux_ has quit IRC06:36
*** acrux has joined #crux06:38
*** jdolan has quit IRC06:56
*** jdolan has joined #crux07:56
*** ChanServ sets mode: +o jdolan07:56
*** DaViruz_ has quit IRC08:01
*** sepen has joined #crux08:07
jaegerMy 2x Xeon is generally much slower than my i7, for what that's worth08:13
jaeger16 threads @ 2.26GHz vs. 8 threads at 3.40GHz08:14
*** DaViruz has joined #crux08:15
teK_my i7 has 12 MB cache and about the same flags as the Xeon + a higher clock.. mhmm08:20
jaeger8M cache on both of mine and similar flags08:23
jaegerclock speed is the biggest difference for me08:23
teK_amd vs. intel has shown that clock is not everything08:29
jaegerof course it's not always but in this case it very much is the difference08:30
jaegerthis is a specific case if that wasn't obvious from the delimiters :)08:30
teK_it is. This was meant as a general comment/"I wonder why"08:40
jaegerfair enough08:41
teK_php screw up  ahead, again :>08:44
*** SiFuh has quit IRC09:34
*** lasso|qt has quit IRC09:36
*** SiFuh has joined #crux09:40
*** lasso|qt has joined #crux09:46
*** lasso|qt has quit IRC11:42
*** nogagplz has quit IRC11:43
*** nogagplz_ has quit IRC11:45
*** tilman has quit IRC12:00
*** tilman has joined #crux12:02
*** ChanServ sets mode: +o tilman12:02
Amnesiafolks, would it be an idea to support binary packages for crux?12:12
Amnesiaif everyone would run "pkgexport $(pkginfo -i)" then it'd be a piece of cake12:12
Amnesiaor is that entirely against the crux philosophy ?12:14
*** fireglow has left #crux12:19
mike_k_Amnesia: CRUX is very relaxed at the package deps and versions by it's nature. would be hard to reliably use someone's packages mixed with self-build.12:23
Amnesiamike_k_: well, I'm mainly talking 'bout core, opt and xorg12:24
mike_k_there is pkg-get, which saves me some time on remote upgrades. but I am installing only from one source on similary configured machine with almost the same set of packages12:25
mike_k_opt is huge enough to mess things up, I guess12:26
Amnesiathink you meant prt-get ?12:27
mike_k_core and xorg do not worth that - little time benefits on reasonable hardware.12:27
mike_k_nope, pkg-get12:27
mike_k_it is almost forgotten tool these days )12:28
mike_k_maybe, Arch linux is the way to go for binary packages with not-so-much bloat12:30
Amnesiamike_k_: neat, but got any mirror?12:30
Amnesiameh don't agree12:30
Amnesiaarch linux still contains quite a lot of bloat12:30
Amnesiaand pkg-utils are also12:31
mike_k_Amnesia: no public mirrors I know of12:31
AmnesiaI know ente had one, but he took it down12:32
*** aarch_ has joined #crux12:33
*** aarch has quit IRC12:37
*** SiFuh has quit IRC12:37
*** krue has quit IRC12:37
*** poulecaca has quit IRC12:38
*** eintopf has quit IRC12:38
*** teK_ has quit IRC12:38
*** s44 has quit IRC12:38
*** lennart has quit IRC12:38
*** acrux has quit IRC12:38
*** vaddi has quit IRC12:38
*** mavrick61 has quit IRC12:38
*** nthwyatt has quit IRC12:38
*** pedja has quit IRC12:38
*** frinnst has quit IRC12:38
*** niklaswe has quit IRC12:38
*** sh4rm4 has quit IRC12:38
*** eighty4 has quit IRC12:38
*** rmull has quit IRC12:38
*** DaViruz has quit IRC12:38
*** sepen has quit IRC12:38
*** joacim has quit IRC12:38
*** rmull_ has joined #crux12:38
*** acrux has joined #crux12:38
*** vaddi has joined #crux12:38
*** mavrick61 has joined #crux12:38
*** nthwyatt has joined #crux12:38
*** pedja has joined #crux12:38
*** frinnst has joined #crux12:38
*** niklaswe has joined #crux12:38
*** poulecaca has joined #crux12:38
*** eintopf has joined #crux12:38
*** DaViruz has joined #crux12:39
*** sepen has joined #crux12:39
*** joacim has joined #crux12:39
*** SiFuh has joined #crux12:40
*** krue has joined #crux12:40
*** sh4rm4 has joined #crux12:40
*** s44 has joined #crux12:40
*** lennart has joined #crux12:40
*** teK_ has joined #crux12:40
*** rmull_ has quit IRC12:44
*** acrux has quit IRC12:44
*** vaddi has quit IRC12:44
*** mavrick61 has quit IRC12:44
*** nthwyatt has quit IRC12:44
*** pedja has quit IRC12:44
*** frinnst has quit IRC12:44
*** niklaswe has quit IRC12:45
*** rmull_ has joined #crux12:45
*** acrux has joined #crux12:45
*** vaddi has joined #crux12:45
*** mavrick61 has joined #crux12:45
*** nthwyatt has joined #crux12:45
*** pedja has joined #crux12:45
*** frinnst has joined #crux12:45
*** niklaswe has joined #crux12:45
*** v33 has joined #crux12:56
AmnesiaRomster: is there any specific reason you didn't add xorg-xkeyboard, xorg-kbcomp, xorg-auth, xorg-xf86-drivers-evdev, xorg-xf86-drivers-synaptics as dependencies for xorg-server?13:00
jaegerare any of them actually requirements?13:06
jaegersynaptics certainly isn't13:06
Amnesiawell, evdev at least is13:06
Amnesiaand afaik without synaptics most mouses won't work either (might be wrong 'bout this subject though)13:07
Amnesiaonly tried it in a vm so far13:07
Amnesiahm it doesn't seem to be a dependency for xorg13:08
AmnesiaI'll install that, will get back on this matter later on13:08
jaegerevdev is a dep of xorg if not xorg-server13:09
*** eighty4_ has joined #crux13:16
Amnesiayes, my apologies13:26
Amnesiajust saw it13:26
Amnesiajaeger: I like the fact that only the bare essentials are installed :)13:44
jaegerwe do try to keep it trim. :)13:47
* Amnesia likes13:50
AmnesiaI also practice the same sport on debian:)13:51
*** cruxbot has quit IRC13:53
* jaeger waves to crux.nu13:53
*** mavrick61 has quit IRC13:54
Amnesiayeah it's off13:56
Amnesiais 2.8 going final :)?13:57
jaegerwhen cruxbot and mavrick61 go at the same time, something went wrong13:57
jaegernot yet, no. we're waiting for gcc 4.7.2 to be released, at least13:57
Amnesiaah ook13:58
*** joacim has quit IRC14:02
v33will there be a difference in performance?14:13
v33i actually did look at the logs romster and  frinnst :)14:14
*** sepen has quit IRC14:33
*** Romeo- has joined #crux14:52
*** lasso has joined #crux14:56
*** nogagplz has joined #crux15:19
*** nogagplz_ has joined #crux15:19
Amnesiahm could anyone over here tell me what the link is between the kernel's graphics drivers and xorg's graphics drivers?15:20
Amnesiahardware -> kernel driver -> xorg driver -> xorg15:20
Amnesia^ is that flowchart correct?15:20
*** mike_k_ has quit IRC15:24
tilmanAmnesia: pretty much15:36
Amnesiaand what's the link between the kernel driver and the xorg driver?15:37
*** vaddi has quit IRC15:38
*** Romeo- has quit IRC15:42
*** Romeo- has joined #crux15:42
*** lasso has quit IRC16:17
*** sammi` has quit IRC16:17
*** sammi` has joined #crux16:28
*** jdolan has quit IRC16:34
*** smartmic has joined #crux17:04
*** tj_ has quit IRC17:04
*** tj_ has joined #crux17:14
*** smartmic has left #crux17:16
*** smartmic has joined #crux17:30
joe9gtk build fails with this message
joe9prt-get info gtk
joe9prt-get fsearch pango-1.0.gir does not return anything17:32
smartmicHi! Today I stumbled on CRUX and I want to try it out... look very promising to me. Still, I have a simply question.17:35
smartmicI never build a kernel by myself and just want to know the time effort to do so. Anybody?17:38
entehow long exactly it takes depends on your hardware17:42
enteplease note that you don't need to compile your own kernel if you don't feel up to it (it's really not that hard), but you can e.g. use the slackware-huge kernel in the beginning17:42
smartmicyes, of course, but also the make config seems a long process (actually, I want to compile one)17:43
entewell, I used slackware-huge in the beginning before I had the time to compile my own17:44
entethe kernel on the crux cd is not such a good idea, because not all hardware is supported17:44
smartmicmy hardware is a qemu x86 emulation on an Corei5 laptop17:44
enteon my p4, it took about 45 min17:44
entemight be faster for you17:45
enteregarding compilation time, the linux kernel is rather harmless17:45
smartmichm, sounds reasonable17:45
Romsteri5 probably do it in 15 minutes with make -j417:45
Romsterlspci -k17:45
Romsterlook for the modules required in make menuconfig17:45
entecompiling gcc on the other hand...17:46
nogagplzbe a man, make -j17:46
Romsteri haven't got timeto run though the entire process17:46
* Romster slaps nogagplz 17:46
* nogagplz cries17:46
ente[00:45] < Romster> i5 probably do it in 15 minutes with make -j417:46
entewell, assuming the i5 is a quadcore, is -j4 not a bit small?17:46
Romsternot for recent kernels there is no BKL now.17:47
nogagplzit's not that I know of, only 2 cores and hyperthreading17:47
smartmicwell, so you recommend menuconfig...are there sensible defaults or do I have to figure them out by myself?17:47
Romsteroh but it might have hyperthreading too.17:47
enteRomster: hm... how is BKL related to that?17:47
smartmicyes 2 core + hyperthreading17:47
smartmicwhat is BKL?17:48
RomsterBKL blocked threads so having more threads was an advantage to not having to wait. now if oyu use too many jobs you'll only slow down the build process.17:48
entesmartmic: big kernel lock17:48
Romsterbig kernel locked since removedfrom the kernel.17:48
enteRomster: but: compilation is I/O bound17:48
Romster-j4 for a i5 that's 2 cores plus hyper threading so you see 4 logical cores17:49
entewhile some gcc has to wait for the hard disk, some other can already parse some file that's already in memory17:49
smartmicI will setup a virtual guest system and want it lean...will have to check qemu docu concerning emulated hardware though17:51
Romsterente, all i know is i'm using BFS and it mentions this
RomsterFor years we've been doing our workloads on linux to have more work than we17:52
Romsterhad CPUs because we thought that the "jobservers" were limited in their17:52
Romsterability to utilise the CPUs effectively (so we did make -j6 or more on a17:52
Romsterquad core machine for example). This scheduler proves that the jobservers17:52
Romsterweren't at fault at all, because make -j4 on a quad core machine with BFS17:52
Romsteris faster than *any* choice of job numbers on CFS.17:52
Romsteroops stupid line breaks.17:52
Romstergoing to work now late.17:53
entenever tried BFS, mostly because it wasn't available for my kernel version when I wanted to try it17:54
smartmicanyhow, thanks for your feedback, I will compile tomorrow or on the weekend...17:56
*** smartmic has quit IRC18:05
*** jdolan has joined #crux18:13
*** ChanServ sets mode: +o jdolan18:13
*** v33 has quit IRC18:25
*** v33 has joined #crux18:33
v33god damn netflix18:37
*** spider44 has joined #crux19:01
*** s44 has quit IRC19:04
*** v33 has quit IRC19:34
*** joacim has joined #crux19:43
*** c0x has quit IRC21:00
*** v33 has joined #crux21:27
*** mavrick61 has joined #crux21:50
*** Amnesia has quit IRC23:46
*** Amnesia has joined #crux23:46
*** vaddi has joined #crux23:53

Generated by 2.11.0 by Marius Gedminas - find it at!